Posts from May 2012 (43)

Stuff is reporting a mind-boggling survey result

Nearly 15 per cent of people worldwide believe the world will end during their lifetime and 10 per cent think the Mayan calendar could signify it will happen in 2012, according to a new poll.

The obvious expectation is that this is a Bogus Poll and that nothing of the sort is true.  However, as the story says, this is a poll conducted for Reuters  by Ipsos, and the findings as given by Ipsos are just as Stuff reports them.

Presumably Ipsos, who know how to poll, are accurately reporting what people said, and the 15% and 10% figures are actually reasonably representative of what people will answer if you ask them that question.  That doesn’t mean the conclusions are true — they obviously aren’t true, since if 15% of people really believed that, they would be behaving differently.

This is a big problem with surveys — even representative polls of people can give weird results, because polls measure how people answer questions, not what they actually believe.

One of my favorite examples is a poll (via) conducted shortly after the massive  Gulf of Mexico oil spill, which found that  “28 percent of Republicans said the recent oil spill in the Gulf of Mexico made them more likely to support drilling off the coast“. This just makes no sense: you could rationally believe that the oil spill is just part of the costs of economic growth and that it doesn’t change your opinion, but supporting drilling more, because of a drilling accident that turned out to be worse and harder to fix than expected, is insane.

Thanks to Kottke‘s excellent blog, I’ve just discovered chartsnthings – a reasonably new blog which details the thinking and work behind some of the New York Times graphs and infographics.

Nice to see they use R (originally developed here at the Department of Statistics at The University of Auckland) to create their graphs.

If you have a research finding that is statistically borderline but makes sense biologically, it often seems that the biological explanation should make it more convincing.  Unfortunately, people are really good at explanations and almost any set of flimsy starting materials can be woven together into a superficially firm explanatory facade.  Keith Baggerley, who analyses gene expression data, says that the cancer researchers he works with can always find explanations for a list of genes that are overactive in a particular experiment — and he’s verified this by giving them completely random gene lists.

The Herald has a nice example in an article on functional foods:

You might be surprised to learn that you can often visually identify the function of natural foods. …Walnuts are nicknamed “brain food” and they look just like tiny brains complete with left and right hemispheres.

Sliced tomatoes resemble the structure of the heart with multiple chambers and, you guessed it, are credited with reducing the risk of heart disease. The list goes on – avocados for the uterus, celery for bones. It seems functional foods have always existed if you look closely.

It helps if you are allowed to stretch both the conditions for similarity and for functionality — celery isn’t distinctively good for bones, and the similarity of avocado to the uterus depends enormously on which variety you choose (or perhaps the Reed cultivar, in season now, and the seedless cocktail avocados are supposed to have other functions).     It also helps if you can ignore badly-fitting examples:  a Canadian paper recently had a similarly credulous story about mushrooms as a functional food, but edible mushrooms are of many different shapes, and the common cultivated mushroom shares its shape with things that you really don’t want to eat.

I won’t even comment on what organ the carrot’s shape shows it supposed to benefit.